PY-L4-01 · JSON Files — Read & Write

Learning Goals

3 min

By the end of this lesson you can:

Explain what JSON is and why it's everywhere on the web.
Read a JSON file into Python with json.load().
Write a Python value to a JSON file with json.dump().
Convert between JSON strings and Python objects using loads / dumps.
Pretty-print JSON so a human can actually read it.

Warm-Up · What Is JSON?

5 min

JSON stands for JavaScript Object Notation. It's a text format that looks almost exactly like a Python dict. Read this file:

{
  "name": "Aisyah",
  "age": 13,
  "subjects": ["maths", "art", "coding"],
  "is_member": true
}

Notice three differences from Python:

Keys are always strings in double quotes — no single quotes, no bare names.
true, false and null are lowercase (vs. Python's True, False, None).
No trailing commas.

Today's big idea

JSON is the universal "data on the wire" format. Every web API, every config file, every save game uses it. Python ships with a json module that turns JSON text into dicts/lists — and back.

New Concept · json.load and json.dump

14 min

Four functions, one rule

The json module gives you four functions. Remember the rule: "s" means string.

json.load(file)    → Python value   (read from a file)
json.dump(value, file) → None         (write to a file)
json.loads(text)   → Python value     (read from a string)
json.dumps(value)  → JSON string      (write to a string)

Reading a file

import json

with open("student.json") as f:
    student = json.load(f)

print(student)
print(student["name"])
print(student["subjects"][0])

{'name': 'Aisyah', 'age': 13, 'subjects': ['maths', 'art', 'coding'], 'is_member': True}
Aisyah
maths

The JSON file becomes a regular Python dict. true in the file became True in Python — the module handles the type translation for you.

Writing a file

import json

scores = {"alice": 90, "bob": 75, "cher": 88}

with open("scores.json", "w") as f:
    json.dump(scores, f, indent=2)

The indent=2 kwarg pretty-prints with two spaces. Without it, you get one cramped line — fine for machines, painful for humans.

Strings instead of files (loads / dumps)

Often the JSON arrives as a string (from an API, from a clipboard paste). Use the "s" versions:

raw = '{"name": "Bob", "age": 12}'
person = json.loads(raw)
print(person["name"])         # → Bob

back_to_text = json.dumps(person)
print(back_to_text)           # → {"name": "Bob", "age": 12}

What types does JSON understand?

JSON         ↔  Python
"text"       ↔  str
42, 3.14     ↔  int, float
true/false   ↔  True/False
null         ↔  None
[...]        ↔  list
{...}        ↔  dict

That's the whole list. Sets, tuples and custom classes are NOT in there — json.dump will raise TypeError if you try. (You'd convert them to lists/dicts first.)

Worked Example · Save and Reload a Quiz

12 min

We'll build a tiny program that saves quiz results to disk, then reloads them next time it runs. This is how every score-saving app on Earth works.

# quiz_save.py — save and reload a quiz result
import json

FILE = "results.json"

# A run of the quiz
result = {
    "player": "Aisyah",
    "score": 8,
    "total": 10,
    "answers": ["A", "C", "B", "B", "D", "A", "C", "C", "B", "A"],
}

# Save
with open(FILE, "w") as f:
    json.dump(result, f, indent=2)

print(f"Saved to {FILE}")

# Reload (pretend a different run of the program)
with open(FILE) as f:
    loaded = json.load(f)

print(f"\nReloaded: {loaded['player']} scored {loaded['score']}/{loaded['total']}")
print("First three answers:", loaded["answers"][:3])

Sample output

Saved to results.json

Reloaded: Aisyah scored 8/10
First three answers: ['A', 'C', 'B']

Open results.json in your editor and you'll see a clean, indented file you could email to a teacher:

{
  "player": "Aisyah",
  "score": 8,
  "total": 10,
  "answers": [
    "A", "C", "B", "B", "D", "A", "C", "C", "B", "A"
  ]
}

Why with open matters

The with block makes sure the file gets closed even if your code crashes mid-write. Forgetting to close a file can leave it half-written and corrupted. Make with a habit.

Try It Yourself

13 min

01 🟢 Save a profile

Build a dict with your name, age and three hobbies. Save it to me.json with indent=2. Open the file in your editor and check it looks neat.

Hint

import json
me = {"name": "Ali", "age": 12, "hobbies": ["chess", "coding", "kuih"]}
with open("me.json", "w") as f:
    json.dump(me, f, indent=2)

02 🟡 Reload and print

Write a separate script that reads me.json back, then prints a friendly one-line summary: Ali, 12, likes chess + coding + kuih.

Hint

import json
with open("me.json") as f:
    me = json.load(f)
print(f"{me['name']}, {me['age']}, likes {' + '.join(me['hobbies'])}.")

03 🔴 dumps round-trip

Take any nested dict and prove a round-trip works: dump to a string, load it back, confirm the result == the original.

Hint

import json
data = {"team": "Hornets", "players": [{"n": "A", "hp": 10}, {"n": "B", "hp": 8}]}
text = json.dumps(data)
back = json.loads(text)
print(data == back)   # → True

JSON round-trips are how programs send data over networks and read it on the other side.

Mini-Challenge · High-Score Hall of Fame

8 min

Build scores_app.py. On each run:

Try to load existing scores from scores.json. If the file doesn't exist yet, start with an empty list.
Ask the user for a name and a score (use input).
Append a new dict {"name": ..., "score": ...}.
Save the list back to the file.
Print the top 3 scores, highest first.

Show one possible solution

# scores_app.py — file-backed leaderboard
import json
from pathlib import Path

FILE = Path("scores.json")

# 1. Load or start empty
if FILE.exists():
    with FILE.open() as f:
        scores = json.load(f)
else:
    scores = []

# 2. Ask user
name  = input("Name: ").strip()
score = int(input("Score: "))

# 3. Append
scores.append({"name": name, "score": score})

# 4. Save
with FILE.open("w") as f:
    json.dump(scores, f, indent=2)

# 5. Top 3
top = sorted(scores, key=lambda s: s["score"], reverse=True)[:3]
print("\n🏆 Hall of Fame")
for i, s in enumerate(top, 1):
    print(f"  {i}. {s['name']:<10} {s['score']}")

Non-negotiables: handle the "file doesn't exist yet" case, save with indent, sort by score descending, print top 3. Path.exists() from pathlib is the cleanest way to check.

Recap

3 min

JSON is the universal data-on-disk and data-on-the-wire format. Python's json module gives you four functions: load and dump work with files; loads and dumps work with strings. JSON's six types map cleanly to Python — anything else (sets, tuples, custom classes) you convert first. Always pretty-print with indent= when a human will read the file.

Vocabulary Card

JSON: Plain-text data format. Looks like a Python dict but with stricter rules.
json.load(f): Read a JSON file and return a Python value.
json.dump(v, f): Write a Python value to a JSON file. Use indent=2 for human-readable output.
loads / dumps: Same idea, but the source/sink is a string instead of a file.

Homework

4 min

Build contacts.py. It should:

Load contacts from contacts.json, or start with three sample contacts if the file doesn't exist.
Print every contact in a numbered list.
Let the user add a new contact (name, phone, email).
Save the updated list back to contacts.json with indent=2.

Sample · contacts.py

# contacts.py — JSON-backed contact list
import json
from pathlib import Path

FILE = Path("contacts.json")
SEED = [
    {"name": "Ahmad",  "phone": "012-3000111", "email": "ahmad@example.com"},
    {"name": "Mei",    "phone": "012-7411233", "email": "mei@example.com"},
    {"name": "Suresh", "phone": "017-2289900", "email": "suresh@example.com"},
]

if FILE.exists():
    with FILE.open() as f:
        contacts = json.load(f)
else:
    contacts = SEED

print("📒 Contacts")
for i, c in enumerate(contacts, 1):
    print(f"  {i}. {c['name']:<10} {c['phone']:<14} {c['email']}")

print("\nAdd a new contact:")
new = {
    "name":  input("  name : ").strip(),
    "phone": input("  phone: ").strip(),
    "email": input("  email: ").strip(),
}
contacts.append(new)

with FILE.open("w") as f:
    json.dump(contacts, f, indent=2)

print(f"\nSaved. {len(contacts)} contacts in {FILE}.")

Non-negotiables: graceful start when the file doesn't exist, list printed with numbers, all three fields collected, indent=2 when saving.

json.load(file) → Python value (read from a file) json.dump(value, file) → None (write to a file) json.loads(text) → Python value (read from a string) json.dumps(value) → JSON string (write to a string)

# quiz_save.py — save and reload a quiz result import json FILE = "results.json" # A run of the quiz result = { "player": "Aisyah", "score": 8, "total": 10, "answers": ["A", "C", "B", "B", "D", "A", "C", "C", "B", "A"], } # Save with open(FILE, "w") as f: json.dump(result, f, indent=2) print(f"Saved to {FILE}") # Reload (pretend a different run of the program) with open(FILE) as f: loaded = json.load(f) print(f"\nReloaded: {loaded['player']} scored {loaded['score']}/{loaded['total']}") print("First three answers:", loaded["answers"][:3])

# scores_app.py — file-backed leaderboard import json from pathlib import Path FILE = Path("scores.json") # 1. Load or start empty if FILE.exists(): with FILE.open() as f: scores = json.load(f) else: scores = [] # 2. Ask user name = input("Name: ").strip() score = int(input("Score: ")) # 3. Append scores.append({"name": name, "score": score}) # 4. Save with FILE.open("w") as f: json.dump(scores, f, indent=2) # 5. Top 3 top = sorted(scores, key=lambda s: s["score"], reverse=True)[:3] print("\n🏆 Hall of Fame") for i, s in enumerate(top, 1): print(f" {i}. {s['name']:<10} {s['score']}")